Preventing cardiovascular disease in midlife women with HIV: An examination of facilitators and barriers to heart health behaviors

Abstract

Midlife women with HIV (WWH) are disproportionately impacted by cardiovascular disease (CVD), yet little is known about perceptions of CVD risk and the factors that influence engagement in heart health behaviors in this population. Few if any studies have used a qualitative approach to examine these perceptions, which have important implications for minimizing the negative impact of HIV-related non-communicable diseases, the risk for which increases after midlife. Eighteen midlife WWH (aged 40-59) in Boston, MA completed semi-structured interviews to explore perceptions of CVD, HIV, and barriers and facilitators to healthy lifestyle behaviors. Interviews were analyzed via thematic analysis. Participants viewed heart health as important but were unaware of HIV-associated CVD risk. Facilitators included family and generational influences, social support, and access to resources. Physical symptoms, menopause, mental health challenges, and limited financial resources were barriers. Midlife WWH may benefit from tailored CVD prevention interventions that target their unique motivations and barriers to healthy behaviors.

Introduction

As the number of people living with HIV continues to grow, advances in antiretroviral therapy (ART) have contributed to a decline in HIV-related mortality [1]. However, the marked decrease in HIV-related deaths brings with it an increased number of midlife and older adults (aged 40 and above) navigating the complexities of aging with HIV. Women account for a large percentage of older people living with HIV; indeed, the prevalence of older women (aged 50 and above) with HIV per 100,000 people in the United States (US) grew by 83% from 2007 to 2015 [2, 3], and 28.5% of new HIV diagnoses in 2018 occurred among women 50 and older [4].

Although increased longevity brings hope with respect to HIV-related outcomes, it is also associated with age-related physical challenges and comorbidities, such as cardiovascular disease (CVD), which is contributing to the global burden of non-communicable diseases (NCDs) among women. CVD is twice as likely to develop among people with HIV relative to the general population [5]. A combination of direct HIV effects and the associated dyslipidemia caused by antiretroviral therapies (ART), increased inflammation, irregular coagulation, and monocyte activation are likely major contributors to the risk of CVD among people living with HIV [6]. Moreover, people living with HIV experience disparities in cardiovascular care, such that they are less likely to receive aspirin, antiplatelet therapy, stain therapy, and smoking cessation resources compared to HIV uninfected individuals [7]. Women with HIV (WWH) are at even greater relative risk compared to men with HIV, as certain cardiometabolic risk factors for CVD (e.g., obesity, metabolic syndrome) are more common in women [8]. WWH are also at greater risk for poor cardiovascular outcomes relative to HIV-negative women. For example, women WWH have higher odds [aOR = 2.61] of dyslipidemia [9], which may be due to overproduction of hepatic very low density lipoprotein cholesterol, decreased clearance of triglycerides, or low levels of HDL cholesterol, all of which have been associated with HIV [10, 11]. WWH are also significantly more likely to smoke tobacco than HIV-negative women, and smoking is independently associated with both dyslipidemia and metabolic syndrome [9, 12]. However, the exact mechanisms driving the heightened CVD risk among WWH are still being explored, in part due to the historic underrepresentation of women in CVD research [13, 14]. Contributing factors may include: the effects of (early) menopause and associated estrogen loss [15, 16]; underuse of CVD screening and treatment interventions relative to men [17, 18]; and greater abdominal visceral fat compared to seronegative women [19, 20]. Given the heightened risk for CVD among WWH, it is important to have a strong and nuanced understanding of CVD knowledge and perceived risk for CVD, as established strategies and interventions that facilitate CVD prevention or mitigate risk may need to be adapted for this population.

Lifestyle modification may reduce the risk of CVD among people living with HIV; however, initiating and sustaining healthy lifestyle behaviors is often challenging, particularly for WWH [21]. Only 48% of WWH in the US report any physical activity of moderate intensity in the past week [21]. Furthermore, dropout rates from physical activity interventions are high among people living with HIV relative to other chronic illness populations [22], suggesting the existence of unique obstacles to engaging in heart health behaviors (e.g., low socioeconomic status [SES[23]). Among WWH, barriers to exercise may include: poor physical health; fear of exacerbating existing medical conditions [21]; lack of motivation or low perceived self-efficacy [24]; and depression, which is more prevalent among WWH than men with HIV [25]. Additional barriers to physical activity that may be relevant to midlife WWH include menopause, associated hot flashes, and related fatigue [26, 27], though findings are inconsistent [28] and physical activity may be more limited by health status than by menopause [29]. As such, health and menopausal status were included in a theoretical conceptualization of midlife women’s attitudes toward physical activity, among other factors (e.g., self-efficacy, social influence). Fear of inducing fatigue or muscle soreness and exacerbating existing health conditions or injuries may also deter participation in physical activities [30]. Further, maintaining a heart healthy diet (e.g., reducing saturated fat and cholesterol consumption) may be difficult due to food insecurity [31]; which has been associated with obesity among WWH [32] and other structural barriers common in populations with low SES.

Although the few studies referenced above have quantified some of the barriers to healthy lifestyle behaviors in WWH, none have used qualitative methodologies to examine the perceived relationship between CVD and HIV, current heart health behaviors, associated motivations or facilitators of those behaviors, and barriers to initiating and/or maintaining heart health behaviors among midlife WWH. Existing quantitative data provide breadth of information on these relationships, but qualitative data would offer depth of knowledge around motivations and barriers that is essential for intervention development. Indeed, the absence of qualitative data on heart health in midlife WWH makes it difficult to determine how best to approach preventive care and to sustainably support heart health in this at-risk population. Conceptualization and improved understanding of these factors is needed to mobilize women for behavior change, capitalizing on their existing motivations, and to inform the development and tailoring of effective interventions. Therefore, the present study aimed to explore among midlife WWH: (1) perceived CVD risk and awareness of the relationship between HIV and CVD; (2) current healthy lifestyle behaviors as well as associated motivations and facilitators of those behaviors; and (3) barriers to heart health behaviors and sustained behavior change. Findings informed the development of a healthy lifestyle intervention tailored to the specific needs of midlife WWH.

Methods

Participants and Recruitment

Midlife WWH were purposively sampled using flyers and presentations at HIV-focused community organizations in the greater Boston area. Potentially eligible women were screened via phone by a research assistant. Eligibility criteria included: (1) living with HIV; (2) cisgender women, which was based on a combination of a verbal confirmation of current gender and an assessment of sex assigned at birth (i.e., “Are you a biologically born woman?”) during the phone screen ); (3) aged 40-59; (4) English-speaking; and (5) able to provide informed consent. Participants were excluded if they (1) were unwilling to participate in the audio-taped interview or (2) had an active (i.e., untreated) and interfering severe psychiatric illness (e.g., bipolar disorder, schizophrenia) that would hinder participation in an interview. The second exclusion criterion was assessed among potential participants who had difficulty answering screening questions (i.e., due to apparent disorganized or pressured speech, flight of ideas, or other verbal symptoms of mania or psychosis that could be identified over the phone). These women were transferred to the principal investigator (PI), a licensed clinical psychologist, who conducted an additional assessment by phone, using her clinical judgment to determine if a potential participant would be able to complete a structured interview. The PI discussed local psychiatric care options with women who were deemed unable to complete an interview.

Eligible individuals attended one in-person visit at a Boston-area hospital. Prior to the initiation of study procedures, the principal investigator and/or a trained research assistant conducted the informed consent process, in a private setting, during which the informed consent document was reviewed together with the eligible individual in detail, and opportunities to ask questions were provided throughout the discussion. Potential participants were informed that participation in the study is entirely voluntary, that they would be permitted to drop out at any time without consequence, and that their medical care within the hospital system would be unaffected by their decision to participate. Individuals who agreed to participate were asked to sign the form, and they were then provided with a physical copy of the signed document for their records.

Participants then completed a self-report quantitative assessment, followed by a semi-structured, audio-recorded, individual qualitative interview, which was conducted by either the principal investigator or a trained research assistant. All interviews were conducted in a private conference room within the hospital. Participants were compensated $25 for their time and transportation expenses. Interviews were conducted between November 2017 and June 2018; Institutional Review Board (IRB) approval was obtained from the Partners Human Research Committee (2017P001770).

Measures

Participants self-reported sociodemographic information, including age, race, ethnicity, education level, employment status, income, sexual orientation, relationship status, date of HIV diagnosis, smoking status, and menopause status. Menopause status was assessed by collecting the date of last menstrual period; participants whose last menstrual period was greater than 12 months prior to their assessment were classified as postmenopausal. Data on CVD knowledge, perceived risk of CVD, and quality of life were included to further characterize the sample.

CVD knowledge was measured using the 25-item Heart Disease Fact Questionnaire (HDFQ [33]), a true/false scale with total scores ranging from 0-25 and higher scores indicating greater CVD knowledge. Scores of less than 50% accuracy were classified as “low” knowledge, between 50 and 69% as “moderate” knowledge, and scores at or greater than 70% as “high” level of knowledge [34]. The HDFQ has demonstrated good content and face validity and excellent readability (at or below the level of a 13-year-old). Given the heterogeneous nature of the items and their range of difficulty levels, the HDFQ also has respectable reliability (Kuder-Richardson formula-20 for internal consistency = 0.77) [33].

Perceived risk of CVD was measured via the 20-item Perception of Risk of Heart Disease Scale (PRHDS [35]); items are rated on a 4-point Likert scale, with total scores ranging from 20-80 and higher scores indicating a higher perceived risk. The PRHDS has acceptable reliability, with a Cronbach’s alpha coefficient of 0.80, as well as acceptable validity (face, content, and construct) [35].

The 20-item AIDS Clinical Trials Group Multidimensional Health Status Assessment (ACTG QOL-602 [36]) was used to measure health-related quality of life; subscale scores are transformed to a scale ranging from 0-100, with higher scores indicating better quality of life. Among individuals living with HIV at various disease stages, the ACTG QOL-602 has demonstrated acceptability validity and reliability (with interclass correlation coefficients > 0.70 for all subscales) [37, 38].

All quantitative data were managed in the Research Electronic Data Capture (REDCap) system [39].

Semi-structured Qualitative Interviews.

The interview guide was developed through a review of existing literature and input from study team members. The guide was also informed by the Health Belief Model, which specifies that engagement in health behaviors is based on one’s perceived vulnerability to and severity of health problems, benefits and barriers to action, and self-efficacy [40]. Interview questions were broadly categorized into three domains (See Table 1 for sample questions and probes): perceived CVD risk and awareness of the relationship between HIV and CVD (CVD risk); current healthy lifestyle behaviors as well as associated motivations and facilitators of those behaviors (current heart health behaviors); and barriers to heart health behaviors and sustained behavior change (potential barriers). In the CVD risk section, participants were asked to describe the significance of heart health in their daily lives and consider their relative risk. Within the current heart health behaviors domain, participants were asked to identify current healthy behaviors, describe facilitators of those behaviors, and express their motivations for staying healthy. Finally, participants were asked to identify barriers to healthy behaviors, and they were probed on the following potential barriers: physical health; emotional health; self-efficacy and uncertainty; social support and stigma; and accessibility.

Table 1.

Sample Interview Domains, Questions and Probes

Domains	Sample Questions and Probes
CVD Risk	Have you and your healthcare provider had a conversation about your heart health, and what you can do to maintain or improve your heart health? How do you see your personal risk for developing cardiovascular disease? Why?
Current Heart Health Behaviors	What do you do to keep yourself and your heart healthy? (Participants probed until no additional behaviors are named.) What challenges do you face in doing ______? (Asked for each behavior; participants probed until no additional challenges were named.) What helps or supports you in doing ______? (Asked for each behavior; participants probed until no additional facilitators were named.)
Potential Barriers	Physical Health: Before you mentioned that ______ is/are problem(s) for you. How have health problems impacted your engagement in healthy lifestyle behaviors?
	Emotional Health: Has depression, anxiety, or any other emotional health problem ever been an issue for you? Please tell me about it.
	Self-efficacy / Uncertainty: How much control do you feel you have over your body and physical health?
	Social Support / Stigma: Tell me a little about your support. Who is in your system? What kind of support do you feel is missing? I am also interested in hearing about your experiences with stigma or discrimination. Have you had these kinds of experiences? Tell me about them.
	Accessibility: Please describe your access to healthy lifestyle resources. Do you feel that you have what you need to engage in healthy behaviors? What is missing?
	Asked for each potential barrier: Tell me how ______ has affected your ability to engage in healthy lifestyle behaviors.

Sample content areas, questions, and probes from in-depth, semi-structured qualitative interviews.

Analyses

Descriptive statistics were calculated to characterize the sample, and the interviews were professionally transcribed. Guided by the principles of thematic analysis [41], three study team members first reviewed several interview transcripts independently, and then discussed them as a group, in order to generate an overarching thematic framework for data interpretation. The coding framework (i.e., codebook) was developed iteratively as transcripts were reviewed and major and minor themes were identified. The team members then independently coded each transcript into content areas and, after completing their independent coding, met to re-examine and discuss core themes, compare codes, and resolve discrepancies; they also maintained an audit trail at each stage of analysis. All data were coded and analyzed using NVivo version 12 [42]. After all coding was complete, the coded data were then re-examined and discussed by the team members in order to extract salient messages across key themes.

The credibility of the data was ensured through investigator triangulation [43] and by verifying participants’ experiences during the interviews. Investigator triangulation was applied by involving three researchers in study execution and in the analysis process as coders. The first several interviews were independently coded by all three study team members, and then discussed as a group, in order to ensure consistency of coding. The remaining interviews were then independently coded by at least two of the three coders, who met regularly to continue comparing interpretations and, if those interpretations differed, worked together to resolve the discrepancy. Following Lincoln and Guba’s recommendations [44], we also ensured that the interviewers conducted one or two pilot interviews with each other to refine the process as well as to ensure effective time management during the study. In addition, participants were asked several distinct questions about important topics and were guided to support their comments with specific examples. Interviewers summarized response and reflected them back to the participants, which ensured that incorrect interpretations could be challenged and corrected in the record.

Results

Characteristics of Study Participants

Eighteen participants completed quantitative questionnaires. Sociodemographic data are presented in Table 2. The mean age was 49 years, and more than half of participants self-identified as Black/African American and not Hispanic/Latina. The majority reported receiving disability payments, with an annual income of $10,000 or less. With respect to CVD knowledge, 66.7% of participants (n = 12) had “high” CVD knowledge (i.e., at or greater than 70% accuracy), 27.8% of participants (n = 5) had “moderate” CVD knowledge (i.e., between 50% and 69% accuracy), and 5.6% (n = 1) had “low” CVD knowledge (i.e., less than 50% accuracy). Participants exhibited moderate overall perceived risk of CVD (total score: M = 52.56, SD = 8.01). On health-related quality of life, participants scored lowest on the energy/fatigue domain (M = 43.89) and highest on the role functioning domain (M = 73.51).

Table 2.

Sociodemographic Characteristics and Descriptive Statistics (N = 18)

	n	%
Age (in years)
Mean (SD)	49.1 (5.5)	–
Race
Black/African American	10	55.6
White	6	33.3
More than one race	1	5.6
Unknown or unreported race	1	5.6
Ethnicity
Hispanic or Latina	2	11.1
Not Hispanic or Latina	14	77.8
Unknown or unreported ethnicity	2	11.1
Education
Less than high school degree	4	22.2
High school graduate/GED	5	27.8
Some college or college graduate	8	44.4
Some graduate education	1	5.6
Employment
On disability	14	77.8
Full-time	3	16.7
Part-time	1	5.6
Income
$10,000 or less	12	66.7
$10,001 to $20,000	2	11.1
$20,001 to $40,000	3	16.7
$40,001 to $60,000	–	–
$60,001 to $80,000	1	5.6
Other/prefer not to answer	–	–
Sexual orientation
Exclusively heterosexual	15	83.3
Heterosexual with some homosexual experience	2	11.1
Bisexual	1	5.6
Relationship status
Single	8	44.4
In committed relationship or domestic partnership	2	11.1
Married	3	16.7
Separated or divorced	4	22.2
Widowed	1	5.6
Smoking status
Current	6	33.3
Former	4	22.2
Never	8	44.5
Menopause status
Post-menopausal	8*	44.4
	M (SD)	Possible Range (Actual Range)
CVD knowledge (HDFQ)1	18.11 (4.75)	0-25 (5-25)
CVD risk perception (PRHDS)2
Total score	52.56 (8.01)	20-80 (26-63)
Dread risk	16.44 (3.73)	7-28 (7-21)
Risk	16.44 (3.94)	6-24 (6-21)
Unknown risk	16.56 (2.87)	6-24 (12-22)
Health-related quality of life (ACTG QOL-602)3
Physical functioning	68.75 (28.2)	0-100 (12.5-100)
Role functioning	73.61 (31.47)	0-100 (0-100)
Social functioning	67.28 (34.61)	0-100 (0-99.99)
Cognitive functioning	63.27 (29.89)	0-100 (13.32-99.9)
Pain	62.96 (31.65)	0-100 (0-99.99)
Mental health	57.35 (22.51)	0-100 (6.66-99.9)
Energy/fatigue	43.89 (25.93)	0-100 (0-100)

Sociodemographic data and descriptive statistics calculated for the three quantitative variables (CVD knowledge, perceived risk of CVD, and health-related quality of life). Percentages may not total 100 due to rounding.

^*Data from two participants were invalid and therefore not included.

¹HDFQ: Higher scores indicate greater CVD knowledge.

²PRHDS: Higher scores indicate a higher perceived risk of CVD.

³ACTG QOL-602: Higher scores indicate better health-related quality of life.

Qualitative Findings

Eighteen participants completed semi-structured interviews, and the average length of the interviews was 60 minutes. Participants’ responses to the probes outlined in Table 1 were organized into three domains, based on the research questions, and specific themes emerged from these domains.

Domain 1: Perceived CVD risk and awareness of the relationship between HIV and CVD

Theme 1a: High perceived importance of and high perceived risk for CVD.

Though three participants had low perceived risk of CVD, most of the women in the sample reported that they considered themselves to be at relatively high risk for cardiovascular problems, primarily due to family history and age. Relatedly, many women described heart health as being of high importance to them. One participant expressed a high level of concern about CVD risk given her extensive family history:

“Heart disease runs on the female side of my family. My grandmother died of a heart attack in her sleep. So I’m very concerned. And my aunt on my mother’s side has a bad heart.”

(52 years old)

Another participant noted that her age increases the salience of CVD risk:

“Now since I'm older… I'm aware of it more now than I was before.”

(54 years old)

Theme 1b: Little knowledge of HIV/CVD connection.

Though participants generally indicated high perceived risk for CVD and considered CVD to be of relatively high importance, they were mostly unaware of the relationship between HIV and CVD. One woman reported “zero” perceived risk for CVD, unaware of the additional risk conferred by her HIV status. Indeed, this participant had the lowest risk perception score of all participants in the study (26; see Table 2 for range). When asked why she reported “zero” perceived risk for CVD, she explained:

“Because it doesn’t run in my family… And I’m not eating like a whole lot of fatty foods.”

(40 years old)

Other participants verbalized that the topic of the study led them to realize that HIV might be associated increased CVD risk, revealing that they otherwise had no information about the CVD-HIV connection. One participant expressed this sentiment:

“Just from what I've heard from this conversation so far, and you're doing a study about how HIV could also be a contributing factor for me getting heart disease.”

(54 years old)

Similarly, another participant shared that this interview was the first time that she was learning about the possible negative implications of HIV for heart health:

“This is the first I’ve ever heard that it [HIV] could impact your heart.”

(51 years old)

Domain 2: Current healthy lifestyle behaviors as well as associated motivations and facilitators of those behaviors

Theme 2a: Eating “right”, reducing alcohol use, and physical activity to promote health.

Participants described a range of current healthy lifestyle behaviors, including healthy eating, portion control, and reduced alcohol consumption. One participant noted that she has been focusing on consuming smaller portions as well as more fruits and vegetables:

“I've been practicing eating small portions, and I've been trying to eat more fruits and vegetables, and you know, like nuts and stuff like that.”

(59 years old)

Another participant indicated that she has made efforts to reduce her alcohol consumption. When she successfully does so, she explained that she rewards herself:

“You see the pink bows there [on my calendar]? Those are my days that I haven’t drank. And every day… I put a pink bow on my calendar, just to let me know that this is the day that was good, and… I did what I was supposed to do for that day.”

(50 years old)

Several women recognized the importance of engaging in physical activity and described planned walks and other behaviors that keep the body in motion. One participant expressed this sentiment, noting that physical activity is also weight management strategy:

“I try to stay active. If I keep my body moving, then I’m OK. Like if I wasn’t as active as I am, I think I will probably be like 800 pounds.”

(42 years old)

In addition to focusing on healthy eating, reduced alcohol use, and physical activity, women acknowledged that daily stressors may exacerbate physical health challenges and described current efforts to better manage stress. One participant reported that she is more intentionally encouraging herself to let go of certain stressors and engaging in adaptive self-talk so as to avoid negative HIV-related outcomes:

“I don't want the HIV to replicate more because I've stressed myself out to the limit. It's not worth all that, so… I need to choose what I'm going to stress myself over and try to learn, as my mother used to tell me, ‘Baby, just let it go.’”

(54 years old)

Theme 2b: Family and generational influences motivate healthy behaviors.

Women described and provided examples of personal experiences with parents who coped with diabetes and other chronic health issues, and they indicated that these experiences motivated them to modify their behavior. One participant reported that her father’s extensive health history encouraged her to act:

“With my father’s history and stuff… And, you know, being a diabetic, having him have the bypass surgery, it kind of… triggered something like, ‘You got to do something.’ … I mean, I’m HIV-positive. I don’t want to die young. You know? I want to, like, keep living as far – as long as I can.”

(51 years old)

Similarly, another participant expressed the impact of her mother’s health issues on her own behavior, noting that the experience of caretaking for her mother made her ever more aware of the potential negative effects of her health status on her own children:

“I was in a hospital a whole month with [my mom]… I never left her side… And every day I had to do like, pick her up and lift her and change her and – so once my mom passed away, it hit me. Do I really want to put this on my kids?”

(44 years old)

In addition to observing the effects of chronic health conditions on parents and other members of previous generations, some participants explicitly stated that their children and grandchildren act as motivators for behavior change. One participant described the importance of modeling healthy behaviors for her son, indicating that she hopes to act as an example:

“Because [my son is] having health issues as well, and I worry that if I can't get him to change until he's ready to change, at least he could see his mom as being an example of trying to do better health wise, cook better, have better food in the house.”

(54 years old)

Another participant expressed the impact of her granddaughter’s eagerness to engage with her on her decision to adopt healthy lifestyle behaviors. Specifically, she reported that she feels compelled to act so that she can continue to be in her granddaughter’s life:

“Every time her mother talks to me, [my granddaughter] is like, ‘I want to talk… can I say hi?’… So, that’s like, ‘Oh, somebody loves me… somebody wants to be around me’… let me get up and do something so that I can be available for her.”

(55 years old)

Theme 2c: Social support facilitates health behaviors.

Women reported that multiple forms of social support help them maintain their heart health, including from agencies that serve people living with HIV, therapists and caseworkers, and family and friends who hold participants accountable for staying healthy. One participant indicated that a wide network of providers help her manage and control her heart health:

“You get the groups, you get the caseworker, you get the therapist, you get the psychiatrist… If you're in depression and you let the therapist know, like leave her a message, she'll either do a home visit… or she'll have you come in, if you could. And they talk you through it and so you have different methods of controlling it.”

(44 years old)

Another participant specified that the social support to maintain her heart health comes from friends who participate in healthy activities with her and help to hold her accountable:

“The Zumba class I do with a couple of friends… they’re the ones that told me about it. So, you know, if I have people… doing things with me, that also helps, if you can do it in a group – and keep each other accountable.”

(50 years old)

For other women, the support to initiate and maintain healthy behaviors comes from close family members. One participant described her son’s efforts to encourage healthy eating:

“My son, he’s a healthy eater, so if he’s seen me, you know, like when we go shopping… I’ll pick up something, he’ll read the label. ‘Mummy, this got too much sugar in it.’ And he’ll put it back, and I get mad, he was like… ‘Why you getting mad? I’m only looking out for your health.’ You know, so … he wants me to eat more healthy.”

(49 years old)

Theme 2d: Access to resources facilitates health behaviors.

Participants noted that access to resources, like green spaces, gyms, and health education, increases their motivation to maintain heart heath. For one woman, access to a park is crucial:

“I’m in a vicinity where I live right across the street from the park, so there’s, like, a tennis court, a basketball court, and then there’s the big park steps, and you can walk all around… I’m just right across the street.”

(55 years old)

For another participant, access to nurse educators through one of the organizations with which she is affiliated provided information on the relationship between dental and heart health:

“There’s a group of nurses that come to one of the centers that I go to that talk to you about heart health… the other day, they were talking about your dental health because it also helps with your heart… They come in and take your blood pressure and give things you can do to try to help a little bit.”

(40 years old)

Similarly, another participant expressed the value of nutritional classes, noting that they provide information on ingredients that may reduce CVD risk:

“I have access to nutritional classes. Which help me. They talk about heart disease, they talk about what butters and what oils… So that's a good thing.”

(42 years old)

Domain 3: Barriers to heart health behaviors and sustained behavior change

Theme 3a: Physical conditions, menopause, and medication side effects act as barriers.

Physical health conditions, the perceived symptoms of menopause, and medication side effects both related and unrelated to HIV appeared to negatively influence engagement in healthy behaviors and activities. Physical conditions exacerbated by HIV or associated with HIV were noted as barriers. For example, one participant reported that her compromised immune system leaves her sick for long periods of time, making it difficult to draw upon her healthy lifestyle resources during episodes of extended illness:

“Living with HIV, like for me anyway, when I get sick, I get deadly sick… I'm down and out for two weeks, so that means I haven't been to the park for two weeks. I haven't been probably to a meeting. I've been stuck in the house with this cold.”

(48 years old)

Several women described the negative impact of menopause and perceived menopause-related symptoms (e.g., hot flashes, associated sleeplessness, decreased energy) on their physical wellbeing, such that these symptoms hindered their ability to engage in healthy lifestyle behaviors. One participant attributed her less frequent attendance at the gym to the onset and continuation of her menopause symptoms:

“Serious menopause… this middle age stuff. It slows you down; it takes away your energy… I just feel older.”

(55 years old)

In addition, participants reported that medication side effects compromise healthy lifestyle goals. One woman indicated that, though she is aware of the hazards of eating “junk food,” she finds that her medications increase her hunger at night:

“I try not to eat junk food before I go to sleep, because that was like [an] influence on me. I mean sometimes my [HIV] medicine will make me hungry and eat, and you know, and I find myself eating before I go to bed.”

(42 years old)

Theme 3b: Mental health challenges act as barriers.

Participants discussed the ways in which depressive symptoms often lead to isolation, unhealthy eating habits, and low levels of physical activity. One participant identified the ways in which her depression affects her ability to engage in several heart health behaviors:

“When I’m depressed, I’m not doing any of those things… I’m not working out. I’m not eating well… I’m also drinking, so, you know, that has a lot to do with it.”

(50 years old)

Menopause was also mentioned in the context of mental health-related barriers to healthy lifestyle behaviors, with one participant attributing her depressive symptoms to menopause:

“It takes away your mental stability, kind of like you become a little bit more depressed. Right now I take antidepressants because of menopause.”

(55 years old)

Several participants reported that lack of motivation and low mood often left them stuck in negative thinking cycles that are hard to break. One woman noted that these cognitive cycles, coupled with her existing health challenges, led her to “give up” on some healthy behaviors:

“Because if you’re not motivated, your body’s going one way, and you want to go this way. It’s like you split in half… And because of my health factors, I can’t do the things. So it puts you in a form of depression. Why do it, if you’re not going to succeed, you know? You just give up. And that’s what I do.”

(49 years old)

For another participant, negative thinking stemming from lack of confidence and poor body image ultimately led to decreased physical activity. This woman reported that, when she feels bad about herself and her body, she starts to believe that she will never find another partner, which then reduces motivation to engage in healthy behaviors:

“I feel less attractive… And because I feel that way, I feel like I’m never going to have another partner. So, because I feel like I’m never going to have another partner, it takes away the motivation to do anything for myself, like to do the exercise.”

(55 years old)

Theme 3c: Limited financial resources as a barrier.

Limited financial resources minimized opportunities to make healthy choices and further compromised participants’ efforts to increase health-oriented behaviors. One participant reported that her income generally precludes healthy eating, noting that, when food is available, she ensures that her son eats first:

“I don’t eat that much… Because I live on $15 of food stamps a month… and my youngest son lives with me. So I make sure he eats before I eat.”

(43 years old)

Another participant indicated that her financial situation limits her ability to engage with the support systems that other women described as facilitators of heart health behaviors. Indeed, her lack of resources forces her to choose among her providers on any given month:

“If I don't have enough money… I have to choose between getting to my doctors, getting to my therapist, and can't go to the other agencies for support, because I have to choose which ones are more important for me for that month.”

(54 years old)

Discussion

In this study, midlife WWH reflected on their own heart health, the relationship between CVD and HIV, current healthy lifestyle behaviors, and barriers and facilitators to engaging in and maintaining additional healthy behaviors. Notable findings include (1) the fact that midlife WWH may be unaware of the increased risk for CVD that is associated with HIV, (2) the influence of family and generational factors (i.e., family history of chronic illness, the desire to serve as a model of healthy living for future generations) on motivation to engage in heart health behaviors, and (3) the possibility that menopause symptoms, including depression, may act as a barrier to heart health activities. Interventions that seek to increase heart health behaviors and reduce CVD risk in this population will likely benefit from leveraging generational influences and including concrete skills that help WWH initiate and maintain these behaviors during periods of physical distress, depression, and pain.

Perceived CVD risk and awareness of the relationship between HIV and CVD

The ways in which midlife WWH conceptualize the relationship between HIV and CVD and describe their motivations for behavior change are critical to explore in the context of intervention adaptation and development. Though most participants considered themselves to be at high risk for CVD and recognized the importance of heart health, they were generally unaware of the additional risk for CVD associated with HIV. This is a novel finding, as there are little available data on CVD risk perception in this population. Compared to a group of mostly men living with HIV (mean age = 48 years), it appears that women in our sample perceived themselves as slightly less vulnerable to CVD (average of 52.6 vs. 53.1 on the PRHDS) and exhibited slightly less CVD knowledge (average of 18.11 vs. 19 on the HDFQ) [45]. Among HIV-uninfected women in the US, CVD knowledge was higher than it was among the midlife WWH in the current sample [46]; in the HIV-uninfected women, knowledge differed by education level and income, consistent with previous reports that CVD knowledge is higher in populations with higher education levels [47, 48] and lower among women with more financial strain [48, 49]. In general, however, studies among seronegative women have demonstrated a broad lack of awareness of their risk for developing CVD [50-53]. Women in the current sample also reported poorer quality of life compared to mostly male samples of people living with HIV [45, 54], and younger WWH [55] in prior studies. These departures may be due, in part, to gender- and/or age-related physiological differences across groups, or the unique socioeconomic challenges in our sample, as the majority of our participants reported annual incomes of less than $10,000. These comparisons suggest that there may be factors contributing to low CVD knowledge and risk perception, as well as to poor health-related quality of life, that are unique to midlife WWH.

It is evident that WWH face challenges assessing their personal CVD risk, suggesting that midlife WWH may benefit from interventions or clinical strategies (e.g., psychoeducation on the additional CVD risk conferred by an HIV diagnosis) that improve CVD risk perception. Relatedly, provider-level interventions may be warranted; if WWH are not aware of the additional risk for CVD that is associated with HIV, their providers may not be highlighting this relationship during medical appointments. Given the growing burden of non-communicable diseases (NCDs) among people living with HIV [56, 57], programs to integrate NCD care into HIV services are being developed across the globe [58-60]. In settings where these programs do not yet exist, providers can be trained to increase CVD knowledge and risk perception by addressing heart health as an important aspect of living well with HIV.

Current healthy lifestyle behaviors as well as associated motivations and facilitators of those behaviors

The specific motivators that catalyze engagement in heart health behaviors among WWH (e.g., to ease children’s future caretaking burdens, to model healthy behaviors), while highly relevant to the design of CVD-focused interventions, are not typically targeted in current programs. Several physical activity-based CVD risk reduction programs have been evaluated [61, 62], but these programs did not leverage or evaluate existing motivations or values to improve or maintain heart health. Relatedly, there are currently no published studies describing psychosocial or behavioral approaches to reducing CVD risk among midlife WWH, though efforts to recruit midlife WWH into CVD prevention studies are underway [63]. As researchers begin to develop CVD risk reduction programs tailored to midlife WWH, they will need to address the unique ways in which this population relates to, defines, and engages with the concept of heart health, particularly given the influence of chronic medical conditions on their relationships with their overall family structures.

Barriers to heart health behaviors and sustained behavior change

Many, but not all, of the barriers to maintaining physical activity and other heart health behaviors discussed by participants were consistent with barriers previously reported by older people living with HIV. Several recent qualitative studies have documented that some older people living with HIV avoid exercise due to fear of injury or concerns about “overdoing it” [64] and low self-efficacy [65], and others described environmental, resource-, and physical health-related constraints as additional barriers to physical activity [66]. Other barriers discussed in this study, including menopause symptoms, have not yet been extensively reviewed elsewhere. Though recent evidence indicates that only vasomotor and urogynecology symptoms may be clearly linked to menopause [67], WWH may consider other symptoms, including decreases in energy or increases in depression symptoms [68], to be associated with menopause and may therefore broadly describe menopause symptoms as barriers to engaging in heart health behaviors. As CVD risk reduction programs are adapted for midlife WWH, menopause symptoms and perceived menopause symptoms will likely need to be considered and addressed, with attention paid to identifying the specific symptoms that are linked to decreased engagement. For instance, cognitive behavioral skills protocols to cope with hot flashes have already been developed [69]. Proactively addressing hot flashes and other menopause symptoms will likely mitigate some of the barriers unique to this population.

Informing the development of CVD prevention programs for midlife WWH

Our sample was generally representative of WWH in the US, suggesting that these motivations and facilitators of engaging in heart health behaviors may be applicable to larger groups of WWH for the purposes of CVD prevention intervention development. For example, the majority of women in our study were Black/African American; in the US, HIV prevalence per 100,000 individuals is 800.9 among Black/African American women, comparted to 45.3 in White women [70]. Poverty likely accounts for some of the racial disparities in HIV prevalence in the US, as 46% of Black/African American women compared to 10% of White women live in poverty [71]. Similarly, most of the participants in this study had annual incomes that placed them below the poverty line in the US, which is currently $12,880 for single-individual households [72]. The implications of low SES and limited resources need to be acknowledged and considered when developing CVD risk reduction program for WLHW. Although research has not thoroughly assessed the ways in which behavioral strategies for CVD management should be adapted to meet the needs of those affected by poverty, some have suggested that the type of interventionist plays an important role. In their review of behavioral strategies for cardiovascular risk reduction in diverse and underserved racial and ethnic groups, Stuart-Shor and colleagues addressed CVD risk reduction in low-income populations [73]. They found that multicomponent, multidisciplinary teams, often led by a nurse, appear to be most effective in addressing multiple risk reduction in these populations, and including culturally and linguistically competent community health workers [74] or peer educators [75] on those teams was an effective strategy. In addition to ensuring that skills and other content included in CVD prevention or risk reduction interventions align with the realities of participants’ financial resources, efforts that target providers, their relationships with communities, and the ways in which they deliver behavioral interventions may be important to achieve sustained improvement [73] among WWH who have low SES.

Though women aged 40-59 do not represent the majority of WWH in the US, this age group would benefit from targeted CVD prevention and/or risk reduction programs. Midlife may actually be an opportune time to engage WWH in behavioral CVD prevention programs, as menopause occurs during this time and the risk of non-communicable diseases increases after midlife. Menopause and its associated estrogen depletion significantly attenuates gender differences in CVD rates [76], and early menopause, which has been associated with HIV [77, 78], is an independent predictor of CVD [79]. In 2011, the Stages and Reproductive Aging Workshop reviewed the findings of cohort studies in the context of chronic illnesses and associated clinical markers among women; from these results, they concluded that the menopausal transition is associated with hormonal and physiological changes [80]. These changes have been linked to an increase in chronic non-communicable conditions after menopause [81], a relationship that is likely exacerbated by poor health behaviors. Therefore, midlife is an appropriate time for WWH to review and potentially alter their health behaviors to support improved health in post-menopause and older age.

With these biological vulnerabilities in mind, a key goal of this study was to inform the development of a CVD-focused intervention for midlife WWH; results from the intervention pilot are reported elsewhere [82], but these findings, though preliminary, may still be applied to the adjustment of existing community-based services for midlife WWH. Services could be expanded to include psychoeducational components that address the behaviors associated with CVD risk, the relationship between CVD and HIV, and the biological factors that exacerbate CVD risk in WWH. Tools for identifying high quality CVD information, increasing social support, accessing local resources, and coping with HIV-related physical health symptoms (e.g., relaxation skills, mindfulness strategies) that interfere with healthy lifestyle behaviors might also be useful. Interventions may also help midlife WWH identify healthy behaviors that align with their values related to caretaking, noting the potential benefits of heart health activities on the health of future generations. The few existing CVD risk reduction interventions that have been developed or modified specifically for people living with HIV [82], such as the HealthMap online self-support management program [83] and the Diabetes Prevention Program [84], can be adapted to address factors relevant to midlife women that were identified in these interviews. Finally, given the possibility of long periods of illness, virtual options for social support and CVD prevention programs should be made available.

Limitations

The current study has several limitations. Participants were recruited from the same health centers and community organizations in the greater Boston area, so their data may not be generalizable to midlife WWH who live in other geographic regions. In addition, we should acknowledge the possibility of researcher bias; because two different researchers conducted the interviews, their approaches to the interviews may have differed. For example, one interviewer may have been more proactive with follow-up questions than the other, or one interviewer may have used more leading questions than the other. Our use of a professional service to transcribe the data may also act as a limitation. When researchers participate in the transcription process, they revisit the interviews and identify possible weaknesses in technique while potentially gaining a deeper understanding of the data; unfortunately, these opportunities are missed when a professional service transcribes the data. Inadvertently, we may have not sampled women who had more knowledge of CVD and/or a strong awareness of the additional CVD risk that is associated with HIV. Our assessment of menopause status may have also been limited. Menopause status was assessed by collecting the date of last menstrual period; participants whose last menstrual period was greater than 12 months prior to their assessment were classified as postmenopausal. Given that we did not plan to assess for differences by menopause status, we did not collect data that would have enabled us to classify the 10 non-postmenopausal women as either premenopausal or peri-menopausal; as such, we are only able to report with certainty that 8 participants were postmenopausal. Finally, while participants’ risk perception related to CVD were measured quantitatively by self-report, current CVD status was not assessed. We did not conduct a medical record review, nor did we collect diagnostic data. Therefore, we were unable to determine if the ways in which facilitators and barriers to healthy lifestyle behaviors differed by CVD status.

Conclusions

Overall, this study demonstrates the ways in which midlife WWH approach heart health and perceive their own risk of CVD, as well as factors that motivate and compromise their efforts to engage in healthy behaviors. Specific concerns that may need to be addressed in future CVD risk reduction interventions for midlife WWH include low CVD risk perception and lack of awareness around HIV as a risk factor for CVD. Reducing CVD risk in WWH requires a nuanced understanding of the (1) perceived relationship between HIV and CVD, (2) existing motivations for healthy behaviors, and (3) challenges that compromise these behaviors. Because WWH have elevated risk for CVD, increased representation of women in HIV-related CVD research is critical to ensuring that this population benefits from advances in the field that are directly relevant to their health and quality of life.

Data Availability Statement:

Due to the nature of this research, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.